Cardiovascular disease (CVD) remains a leading cause of death in the United States. Statin therapy has proven remarkably effective in reducing CVD events when used in those at risk, yet statins remain under- utilized. These gaps in care represent a twofold problem: a failure to appropriately identify candidates for primary prevention and a failure of patients to take these medications when offered. Currently available models to predict CVD risk have significant limitations, particularly for younger adults. Further, their probabilistic outputs are neither easily understood nor impactful to patients. Dr. Navar's K01 proposal uses pooled data from seven NHLBI cohort studies to create an improved CVD risk prediction model for young adults. This model will estimate ?continuous? lifetime CVD risk, incorporating a range of risk factors and evaluating risk at multiple patient ages. Second, Dr. Navar will evaluate how this novel continuous lifetime risk model improves identification of candidates for statin therapy relative to current risk tools. Third, to improve the ability to communicate these data to patients, she will transform these risk estimates into population relative risk. This will allow a patient to understand how his CVD risk profile compares with similar age- and gender-matched peers. She will then use a unique national registry to better understand how best to communicate numerical and probabilistic data to which patients. Finally, she will develop a web-based risk communication tool that incorporates the novel lifetime risk, population relative risk, and current 10-year risk estimates and pilot this tool in patient focus groups. This web-based tool will be linked to the electronic health record and tested in an outpatient clinical setting to assess how real time estimation and patient-tailored shared decision-making affect clinical care. The current proposal will also assist Dr. Navar in fully strengthening her background in public health. The didactic and applied statistical experiences, including training in predictive modeling and causal analysis, will allow her to become more prepared to be a national leader in the understanding of real world safety and effectiveness of preventive therapies. Similarly, the formal training in qualitative research and behavioral science proposed in this application, combined with the practical development of a CVD risk communication tool, will allow her to establish a program of independent research focused on improved uptake of CVD preventive therapies through effective patient risk communication. This training program builds on already successful research at the Duke Clinical Research Institute (DCRI). The mentorship team, led by Dr. Eric Peterson, Director of the DCRI and expert in CVD outcomes research, includes experts in risk prediction (Pencina, co-mentor, and D'Agostino, external advisor) and risk communication (Boulware, co-mentor), all of whom will ensure scientific success and oversee the candidate's advanced training in their relative areas of expertise. By the conclusion of this program, Dr. Navar will be able to carry out advanced research in CVD prevention as an independent physician-scientist.